Lending Club Case Study

Presented by :
Ms. Nazneen Ansari
Mr. Chetan Desai
Batch : ML C36
Date : 9th February, 2022
 Problem Statement:
The lending company facilitates personal loans, business loans, and
financing of medical procedures. Borrowers can easily access lower
interest rate loans through a fast online interface.
However, lending loans to ‘risky’ applicants is the largest source of
financial loss (called credit loss). The credit loss is the amount of money
lost by the lender when the borrower refuses to pay or runs away with
the money owed.
Thus, the company wants to understand the driving factors (or driver
variables) behind loan default, i.e. the variables which are strong
indicators of default. The company can utilise this knowledge for its
portfolio and risk assessment before approving any loan application.

ML C36 Lending Club Case Study 2

Methodology:
• In this case study, EDA will be used to understand how consumer
attributes and loan attributes influence the tendency of default.
• Identifying the two types of risks associated with the loan
applications of the Bank
• If the applicant is likely to repay the loan, then not approving the loan results
in a loss of business to the company
• If the applicant is not likely to repay the loan, i.e. he/she is likely to default,
then approving the loan may lead to a financial loss for the company
• Analysis has to be done on the given data which contains the
information about past loan applicants and whether they ‘defaulted’
or not.
ML C36 Lending Club Case Study 3
Types of Variables
• Quantitative variables :
Rate of Interest, Annual Income, DTI, Funded loan amount, Open to
Total account ratio
• Categorical variables :
• Unordered :
Loan Status, Purpose of Loan, Home Ownership, Verification status
• Ordered :
Year/Month, Term

ML C36 Lending Club Case Study 4

Data Cleaning & Manipulation
• Columns eliminated :
• All the columns with values NA/null
• All the customer behavioural variables
• All the columns with least distinctive values
• Number of columns before data cleaning : 111
• Number of columns after data cleaning : 20
• Outliers have been removed for Annual income
• In Employment length column, the values which are less than 1 year
are taken as 1 year and 10+ are taken as 10 years

ML C36 Lending Club Case Study 5

Correlation
Correlation between attributes

ML C36 Lending Club Case Study 6

Loan Status Distribution

14.97% loans out of total loans are Charged Off

ML C36 Lending Club Case Study 7

 Variable : Verification status

Account holders with annual income between 30k-45k with verification status as Not
Verified tend to default
Account holders who got amount funded by investors between 0-5k and verification status
as Not Verified tend to default
ML C36 Lending Club Case Study 8
 Variable : Term

The ratio of Open-to-Total Account is higher for loans with loan term of 36 months. The
above graph represent that borrowers with more open accounts, taking loan for 36 months
tend to default.
From the above two plots, it is evident that loans with lower term (36 months) and lower
interest rate (~ 12.5%) tend to default comparatively.
ML C36 Lending Club Case Study 9
 Variable :
Home Ownership Vs Funded Amount

People with Mortgage home ownership tend to default with loan amount between 5k-10k,
whereas people with Rent home ownership default with loan amount between 0-5k.

ML C36 Lending Club Case Study 10

 Variable :
Home Ownership Vs Annual Income

People with Annual Income between 30k-45k staying in Rented Accommodation , followed
by people with Annual Income between 45k-60k having Mortgage home ownership.

ML C36 Lending Club Case Study 11

Variable : Employment length Vs Annual Income

Borrowers with 10+ years of work experience and annual income in the range of 45k-60k
tend to default. However, it is also visible that there are considerable number of people
with 1 or less year of work experience and annual income in the range of 15k-45k who
contribute to defaulters list
ML C36 Lending Club Case Study 12
Variable : Employment length Vs Funded Loan Amount

Borrowers with 10+ years of work experience and funded loan amount in the range of 5k-
10k tend to default. Also, people with 1 or less year of work experience and funded loan
amount in the range of 0k-5k contribute to defaulters list.
ML C36 Lending Club Case Study 13
 Variable : States

Maximum loan is borrowed by the people living in California(CA), followed by New York(NY),
Florida(FL), Texas(TX) and New Jersey(NJ). The charged off proportion is also in the same
order of the loan borrowed for these states.

ML C36 Lending Club Case Study 14

 Variable : States

People living in the top 5 states of the U.S., which have the maximum number of Charged
Off loan, borrowed loan for debt consolidation.

ML C36 Lending Club Case Study 15

Correlation –
Loan Purpose Vs Annual Income

The number of loans borrowed for home

improvement & small business with the
annual income around 59k & 58k
respectively contribute to defaulters.

ML C36 Lending Club Case Study 16

Correlation –
Loan Purpose Vs Funded Loan Amount

The number of loans taken for debt

consolidation & small business with
the funded loan amount around 12k
tend to Charged Off.

ML C36 Lending Club Case Study 17

Variable : Grades and Subgrades

From the above two plots, it can be deduced that Loan with grade B and sub-grade B5
comparatively Charged Off more.

ML C36 Lending Club Case Study 18

Variable : Funded Amount Vs Open to Total
account

Charged Off loans are comparatively higher for customers having Open-to-Total Account
Ratio more than 0.5, which means a customer having more Open accounts with funded
loan amount in the range of 0-5k tend to default.

ML C36 Lending Club Case Study 19

Variable :
Annual Income Vs Open to Total account

Charged Off loans are comparatively higher for customers having Open-to-Total Account
Ratio more than 0.6, which means a customer having more Open accounts with annual
income in the range of 0-15k tend to default.

ML C36 Lending Club Case Study 20

 Variable : Debt to Income

The maximum number of Charged Off loans are in the DTI (debt-to-income) range of 12-18.
Loans which are Not Verified in the DTI range of 12-18 tend to Charged Off.

ML C36 Lending Club Case Study 21

Observations & Recommendations
• Loans with funded amount between 0-5k & loans with borrowers
annual income between 30k-45k needs to be verified.
• Borrowers with high open to total account ratio and interest rate (~
12.5%) for the loan term of 36 months, tend to default, hence need
to be monitored
• Borrowers with 10+ years and less than 1 year of work experience
with annual income in the range of 45k-60k and 15k-45k
respectively, and funded amount 5k-10k and 0-5k respectively, tend
to default, hence to be asked to submit some security against the
loan
• Citizens from California who borrowed loan for debt consolidation
need to be watched
ML C36 Lending Club Case Study 22
Observations & Recommendations
• The borrowers with grade B and sub-grade B5 are tend to be Charged
Off more.
• People having rented home tend to charged off, followed by ones
having home ownership as Mortgage.
• Account holders with high open to total account ratio with funded
loan amount of 0-5k and an annual income of 0-15k are more prone
to miss their repayment on time
• The maximum number of Charged Off loans are in the DTI (debt-to-
income) range of 12-18. Loans which are Not Verified in the DTI
range of 12-18 tend to Charged Off.
ML C36 Lending Club Case Study 23
Thank you…!

ML C36 Lending Club Case Study 24